Google Ads Audit: The Complete Checklist to Stop Wasted Spend

A Google Ads audit only earns its time if it changes what you do afterward. The checklists that get shared around usually list 40-plus things to check with no sense of order, which is how an audit turns into a weekend project instead of something you actually finish. This one is organized by how much money is typically leaking through each area, so you check the expensive stuff first and stop wherever your time runs out.

It also includes a category most audit guides fold into a vague "wasted spend" line and move past quickly: invalid clicks and traffic quality. That deserves its own section, not a bullet point, because it's one of the few leaks that keeps growing quietly if nobody checks for it — unlike a bad keyword, which usually shows up in performance data on its own.

Everything here is written to be checked directly inside a standard Google Ads account, with no paid add-on required, so there's no reason a lack of tooling should be the thing that stops this from getting done.

Signs you need one now, not on schedule

Most of the checklist above works fine as a recurring routine — monthly for active accounts, quarterly for stable ones. A few situations move it up the priority list regardless of when the last audit happened.

You took over an account you didn't build. Assume nothing about why anything is configured the way it is until you've checked it yourself; "it's always been set up this way" is not the same as "it's set up correctly," and inherited accounts are where the largest, longest-standing issues tend to live simply because nobody who understood the original setup has looked at it critically in a while.

Performance dropped and you can't immediately explain why. Before assuming it's the market, competition, or seasonality, rule out the boring internal causes first: a conversion action that stopped firing, a budget cap newly limiting a campaign that used to spend freely, or a Quality Score drop on core keywords. Internal causes are more common than external ones and considerably easier to fix once found.

You're about to increase budget significantly. Scaling spend on top of an account with unaddressed structural or tracking issues just scales the waste proportionally. A quick audit before a budget increase is cheap insurance against a much more expensive mistake a month later.

How long this should actually take

Two versions of this audit are worth knowing, because "audit your account" means something different on a $500/month campaign than on a $50,000/month one.

The 20-minute version: conversion tracking sanity check, invalid click rate, search terms report skim for the worst offenders, and budget pacing. This catches the handful of issues that are both common and expensive, and it's worth doing monthly even on small accounts.

The full version below covers eight areas in the order they tend to matter most by dollar impact. On a single, reasonably simple account, budget 90 minutes to 2 hours for a first pass. On an account you didn't build — one you're inheriting, or auditing for a client — budget closer to half a day, because you'll spend real time just understanding why things were set up the way they were before you can tell what's broken versus what's just unfamiliar.

1. Conversion tracking — check this before anything else

Every other number in the account is downstream of this being right. If conversion tracking is broken or double-counting, every optimization decision you make based on it is built on bad data, including the decisions Smart Bidding is making automatically without you.

What to check: open Conversions under Goals and confirm every conversion action you're actively optimizing toward has recorded activity in the last 7 days — a "no recent conversions" warning on a live goal means Smart Bidding has been optimizing blind. Confirm you're not double-counting: a common, easy-to-miss mistake is having both a Google Ads tag and a Google Analytics-imported goal firing for the same action, which inflates your conversion count and quietly makes every campaign look more efficient than it is.

Check that conversion values are populated, not blank, wherever revenue actually varies between conversions — a lead form and a completed $2,000 purchase shouldn't carry equal weight in a bidding algorithm's eyes if they don't carry equal weight in your revenue. And confirm your attribution model was chosen deliberately. Data-driven attribution is the current default for most accounts, but if yours is still on last-click because nobody changed it, that's worth knowing rather than assuming.

One more check worth adding here specifically: if you recently migrated tracking (switching from Universal Analytics to GA4-imported conversions, or moving to server-side tagging), compare conversion volume for a two-week window immediately before and after the migration. A drop of more than about 10% that coincides exactly with the migration date, with no other change to explain it, points at the migration itself rather than genuine demand — a detail that's easy to miss if you're only looking at the current number in isolation.

2. Invalid clicks and traffic quality

This is the section most audit checklists compress into one line about negative keywords. It deserves more room, because it's the one category of waste that doesn't announce itself in a performance report the way a bad keyword does — nothing about the campaign summary tells you a chunk of your clicks never had any chance of converting.

Add "Invalid clicks" and "Invalid click rate" as columns under Campaigns → Modify columns. A rate that's climbing month over month, or a specific campaign running meaningfully hotter than the account average, is worth investigating on its own before you touch anything else in the audit.

Cross-check Google Ads click totals against Google Analytics session totals for the same date range. A meaningful gap — more reported clicks than matching sessions — means some clicks never resulted in a real browser session reaching the site at all, which is a stronger signal than the invalid-click-rate column alone, since that column only shows what Google already caught.

If you find a real pattern here, the fixes are IP exclusions for confirmed sources, tighter geographic targeting if you're seeing traffic from outside your actual service area, and — for anything that looks deliberate or repeat rather than incidental — a dedicated detection layer, since manual IP exclusion runs into Google's 500-address cap fast against anything that rotates.

One thing worth noting explicitly during this check: don't treat a single unusual IP or a short burst of clicks as proof of anything on its own. Shared networks — offices, apartment buildings, mobile carriers using carrier-grade NAT — can put many real, unrelated users behind what looks like one address. The pattern worth acting on is repetition without engagement over time, not a single data point during one audit pass.

3. Account structure

Structure problems are quieter than tracking or fraud issues but compound over time. Look for campaigns effectively competing with each other — the same or near-identical keywords live in two different campaigns, which means you're bidding against your own ads in the auction and paying an inflated price for the privilege.

Check ad group size and theme tightness. An ad group with 40 loosely related keywords produces generic ads that don't match specific search intent well, which drags down Quality Score and raises CPCs across the whole group. Tightly themed ad groups of 5–15 closely related keywords consistently outperform large, loose ones on relevance and cost.

Confirm Search, Shopping, and Performance Max campaigns aren't quietly cannibalizing each other on the same high-value terms — PMax in particular can pull impression share away from a Search campaign targeting the same query without it being obvious unless you're specifically looking for the overlap.

Naming conventions matter more than they seem like they should, mostly because of what they prevent rather than what they enable. A consistent format — location, campaign type, and objective encoded directly in the campaign name — means anyone auditing the account later, including a future version of you, can understand what each campaign is supposed to be doing without opening it. Accounts built without this convention are measurably slower to audit, because half the work becomes reverse-engineering intent before you can even evaluate whether it's working.

4. Keywords, match types, and negative keywords

Pull the search terms report for the last 30–90 days and sort by spend. Anything with meaningful spend and zero conversions is a candidate for a negative keyword, a paused keyword, or a tighter match type — in that rough order of severity.

Check match type distribution. Broad match without Smart Bidding actively managing it, and without a solid negative keyword list underneath it, is the single most common source of irrelevant spend on accounts we see. It's not that broad match is wrong — it can work well with the right bidding strategy and enough negative keyword coverage — it's that broad match with neither is a fast way to burn budget on searches that were never close to your actual offer.

Look for negative keyword conflicts: a positive keyword you're actively bidding on that's also present, or partially matched, inside a negative list somewhere in the account — usually a shared list applied at the account or campaign level. This silently suppresses impressions on a keyword you think is live, and it's genuinely easy to miss because nothing errors out; the keyword just quietly underperforms.

Check for keyword-level duplicate bidding across campaigns too — the same exact-match keyword present in two different campaigns targeting the same geography will compete against itself in the auction, which does nothing but raise your own cost-per-click for no benefit. This is different from intentional overlap (broad match in one campaign, exact match in another, deliberately structured) and worth distinguishing before removing anything.

5. Bidding strategy and budget allocation

Check whether your bid strategy matches your actual goal. Maximize Conversions with no target CPA set will spend your full budget chasing volume, which is fine if volume is genuinely the goal and dangerous if it isn't — it will happily accept a worse cost-per-conversion than you'd tolerate if it means spending the full daily budget.

Review budget allocation against performance, not against how the budgets were originally set. It's common for a campaign that used to be the top performer to keep its original (larger) budget long after a newer campaign has become more efficient, simply because nobody revisited the split.

Check "Search impression share lost (budget)" per campaign. A high number here on a campaign with a good conversion rate means you're leaving profitable volume on the table specifically because the budget caps it, not because demand or targeting is the constraint — that's a straightforward budget-reallocation opportunity, not a strategy problem.

6. Ad copy and assets

Confirm every active ad group has at least two responsive search ads running, so Google has something to actually test against each other — a single ad per group means there's no real optimization happening on copy at all.

Check Ad Strength where it's visible, and more importantly, check whether the copy still reflects your current offer. Stale copy referencing a promotion that ended months ago, or pricing that's since changed, is common on accounts that haven't been touched in a while and is one of the fastest things to fix once spotted.

Confirm sitelinks, callouts, and structured snippets are populated and current. These extensions genuinely move CTR, and an account running without them, or with outdated ones, is leaving a comparatively easy win unclaimed.

7. Audiences and remarketing

Check whether remarketing lists are actually being used anywhere, or whether they exist and sit unattached to any campaign — a surprisingly common finding on accounts that set up remarketing once and never built a campaign around it.

Review audience signals feeding Performance Max, if you're running it, since PMax leans heavily on the audience and customer data you feed it to find similar prospects — a thin or stale audience signal genuinely limits how well it can perform, more than most advertisers realize when setting it up.

Check for audience exclusions where they matter: excluding existing customers from acquisition-focused campaigns, for instance, so you're not paying acquisition-level CPCs to reach people who already converted.

Check remarketing list membership duration against your actual sales cycle. A 30-day remarketing window on a product with a typical 90-day consideration period is excluding genuinely warm prospects from a list built specifically to reach them — a mismatch that's easy to set once during initial setup and then never revisit as the business or the product changes.

8. Landing pages

Confirm the landing page for each major ad group actually matches what the ad promises — a disconnect here (ad promises a specific product or offer, landing page is a generic homepage) hurts Quality Score and conversion rate simultaneously, which is a worse combination than either problem alone.

Check page load speed on mobile specifically, since a meaningful share of search traffic is mobile and a slow-loading page loses visitors before they ever see what you're offering, regardless of how good the ad or the offer is.

Confirm the primary conversion action is easy to complete on the page you're sending traffic to — a form buried below the fold, or a phone number that's an image instead of a clickable tel: link on mobile, are small technical details that measurably suppress conversion rate.

One more landing page check that's easy to overlook: confirm the page still exists and loads correctly at all. Redesigns, CMS migrations, and expired promotional pages occasionally leave an ad group pointing at a 404 or a redirect chain nobody remembers setting up — rare, but the kind of thing that's disproportionately expensive when it does happen, since every click on that ad group is being paid for with nothing to show for it.

Red-flag thresholds worth memorizing

A quick reference for what's worth acting on versus what's normal variation: invalid click rate trending upward over several weeks, more than roughly 5% of spend going to search terms with zero conversions over a 90-day window, any ad group running on a single ad with no second variant to test against, Quality Score below 5 on keywords carrying meaningful spend, and impression share lost to budget above 20% on a campaign that's actually converting well.

None of these are hard rules — context matters, and a brand-new campaign with little data yet will trip several of these harmlessly. They're a starting point for where to look closer, not a verdict on their own.

It's worth revisiting these thresholds every few months rather than treating them as fixed forever — what counts as normal for your account shifts as spend grows, as the account matures past its initial learning phase, and as your market's competitiveness changes. A threshold set when you were spending $500/month may need recalibrating once you're spending $5,000/month, since absolute dollar amounts at a given percentage scale with budget even when the percentage itself doesn't move.

Auditing across a Google Ads manager (MCC) account

If you're running this checklist across several accounts under one MCC, the order changes slightly: run the invalid-click and conversion-tracking checks first, across every account, before doing a deep structural review of any single one. Tracking and fraud issues tend to repeat across accounts that share a setup template or were built by the same person, so catching the pattern early on account one often saves you from re-discovering the same issue independently on accounts two through ten.

Keep a shared running list of what you've checked and when, per account. Without it, it's easy to lose track of which accounts have had a full audit recently and which are overdue, especially once you're managing more than four or five — the accounts that get skipped tend to be whichever ones aren't actively complaining that month, which is exactly the wrong selection criteria.

Common mistakes that undermine an audit

Auditing on too short a data window is the most common one. A 7-day lookback is enough to catch tracking that broke this week, but it's too short to judge keyword or campaign performance fairly, especially for anything with a longer consideration cycle. Use 30–90 days for performance judgments, and reserve the short window for tracking health checks specifically.

Treating every low-converting keyword the same way is another. A keyword with meaningful spend and genuinely zero relevant traffic is a clear pause-or-negative candidate. A keyword with moderate spend and a below-average, but non-zero, conversion rate is a different situation — it might need a bid adjustment or better ad copy rather than removal. Auditing by "anything underperforming average gets cut" tends to over-prune accounts and remove keywords that were contributing, just less efficiently than the top performers.

Fixing everything at once is a mistake that shows up after the audit rather than during it. Changing conversion tracking, bidding strategy, budgets, and ad copy simultaneously means that if performance shifts afterward, good or bad, you have no way to know which change caused it. Sequence changes where you can, especially anything touching bidding strategy or budget, so the account's learning phase isn't reset repeatedly by stacked changes.

Skipping the boring checks because they seem too obvious to be the problem is the last one worth naming. Conversion tracking silently breaking, or a campaign accidentally paused, are unglamorous compared to a fraud pattern or a clever restructuring opportunity — but they're also disproportionately common causes of a sudden performance drop, and checking them takes two minutes. Rule out the boring explanation before spending an hour chasing a more interesting one.

After the audit: what to actually fix first

Fix conversion tracking issues immediately, before anything else, since every other decision in the account depends on that data being accurate — optimizing on top of broken tracking just compounds the problem.

Next, address anything actively losing money right now: invalid traffic patterns, search terms burning spend with zero conversions, and budget caps suppressing a genuinely well-converting campaign. These have the fastest, most measurable payback.

Structural and creative improvements — ad group reorganization, new ad copy, audience refinement — matter, but they're slower-moving and lower-urgency than the first two categories. Doing them first, before tracking and waste are fixed, means you're optimizing a foundation you haven't confirmed is solid yet.

How often should I audit my Google Ads account?

Monthly for the 20-minute version on any actively managed account — conversion tracking, invalid click rate, and a search terms skim. The full audit is worth doing quarterly on a stable account, and immediately whenever you inherit an account, see an unexplained performance drop, or are about to significantly increase budget. Accounts under $1,000/month can usually stretch the full audit to twice a year without much risk, since the dollar impact of most issues scales with spend.

What tools do I need to audit Google Ads?

Everything in this checklist can be done with what's already inside Google Ads and Google Analytics — no paid tool is required to complete a full audit. Google Ads Editor is useful for bulk-reviewing account structure across many campaigns at once rather than clicking through each individually. Third-party audit tools mainly add speed and automated flagging on large or multi-account setups; they don't check anything a manual pass through this list wouldn't eventually find.

How much money does a typical audit actually find?

This varies enormously by how neglected the account has been, so treat any specific percentage you see quoted elsewhere with some skepticism — a well-maintained account audited monthly will find comparatively little each time, while an account that hasn't been reviewed in a year can turn up substantial waste in tracking errors, stale negative keyword gaps, and budget misallocation alone. The honest answer is that the audit's value comes from catching compounding issues early and often, not from one dramatic one-time discovery.

Should I audit Performance Max campaigns differently?

PMax gives you less visibility than Search campaigns by design, so the audit shifts toward inputs rather than direct keyword-level review: audience signal quality, asset group completeness (headlines, descriptions, images, and video), conversion goal configuration feeding the campaign, and the search terms insights report, which is the closest thing PMax offers to a traditional search terms report. You can't negative-keyword your way to precision the way you can in Search, so the input quality matters proportionally more.

Can I audit a competitor's Google Ads the same way?

Not in the same depth — you don't have access to their account, so you're limited to what's publicly visible: their live ad copy via Google's Ad Preview and Diagnosis tool, their approximate positioning via the Auction Insights report on your own overlapping keywords, and general estimates from third-party competitive intelligence tools, which are directionally useful but not precise. A competitor audit answers different questions (positioning, messaging, apparent strategy) than an account audit does (whether your own setup is working correctly).

The one-page version, if you just need the list

Conversion tracking: recent activity on every active goal, no duplicate counting, values populated, attribution model deliberate. Invalid clicks: invalid click rate trend, Ads-vs-Analytics click gap, IP concentration check. Account structure: no internal keyword overlap between campaigns, ad group themes tight, Search/Shopping/PMax not cannibalizing each other. Keywords: search terms report reviewed for zero-conversion spend, match type distribution sane, no negative/positive keyword conflicts.

Bidding and budget: strategy matches actual goal, budget allocation matches current performance not historical assumptions, impression share lost to budget checked on strong performers. Ad copy: 2+ ads per active ad group, copy current and accurate, extensions populated. Audiences: remarketing lists actually attached to campaigns, PMax audience signals current, exclusions in place where relevant. Landing pages: message match, mobile load speed, conversion action easy to complete.

Print or copy that paragraph as a working checklist — it's deliberately dense rather than a a beautifully formatted table, so it's fast to scan the next time you're doing this under time pressure.